Please pass on to any students who might be interested...

*	*	*	*	*	*	*	
Applications are invited from graduates with good honours 
or Masters degrees in relevant disciplines for PhD 
studentships tenable from October 2000. Research areas 
include: quantitative and qualitative health services 
research; aetiological, lifecourse, clinical and 
nutritional epidemiology; effectiveness and acceptability 
of health services/care; health/illness/disability in older 
age.  Projects in statistics/economics/sociology/psychology 
applied to medical/clinical issues also encouraged. At 
least two full-time scholarships are available including 
one from the MRC.  For further details of potential topics 
and supervisors,  and application forms, see below or 
contact: Mrs Julia Mackay/Dr Jenny Donovan, Department of 
Social Medicine, University of Bristol, Canynge Hall, 
Whiteladies Road, Bristol BS8 2PR (0117 928 7337). Email 
[log in to unmask] Web page http://www.epi.bris.ac.uk 
Closing date for applications 30th April 2000. Please note 
that only residents of the EU are eligible for studentships.

*	*	*	*	*	*	*	
Candidates are encouraged to apply to study a topic of 
their own choice in an area related to the Department's 
core interests (see above and http://www.epi.bris.ac.uk), 
or apply for one of the following topics with the named 
adviser(s). You are encouraged to discuss your ideas with 
potential adviser(s). 

1. Prof Shah Ebrahim: Locomotor disability: assessment, 
need and service responses This PhD scholarship is part of 
a programme of work concerned with the assessment, 
prevention and treatment of locomotor disability - a 
problem with walking or balance - among older people. 
Locomotor disability - defined as inability to walk, climb 
stairs or maintain balance - increases almost exponentially 
with age and self-reported locomotor disability affects 
over 3 million people aged 60 years and over (about a third 
of the total) in Britain. Professor Ann Bowling, (UCL), 
Professor John Bond (Newcastle) and Dr Christina Victor (St 
George's Hospital Medical School) are collaborating with 
Professor Shah Ebrahim to commision an Office of National 
Statistics omnibus survey of older people.  This 
cross-sectional survey  of 2000 people will collect data on 
quality of life, social support by informal carers, mood, 
isolation, locomotor disability, activity and participation 
in every day life, use of services, aids and applicances.  
A sub-sample of the participants will be followed up and 
interviewed in depth and all participants will be asked to 
take part in further follow up by postal questionnaire.  We 
seek to answer the following questions: How appropriate and 
valid (from the perspective of older people) are existing 
"standard issue" (as used in censuses) brief disability 
questions?  Does depressed mood affect responses to 
disability questions?  How do quality of life assessments 
and disability measures compare?  How do social and 
economic factors effect levels of disability experienced by 
older people?  Why do older people reject support from 
health and social services?  The data obtained from the ONS 
survey will be used to compare quality of life and 
disability measures, and associations between a wide range 
of social and demographic factors and disability to be 
measured.  Modification of responses to questions about 
disability by depressed mood will be examined.  Comparisons 
between responses to the "standard issue" questions on 
disability and more detailed locomotor disability 
assessment will be examined. Relationships between apparent 
level of need, as assessed by level of disability, and use 
of services, aids and appliances will be studied.  Reasons 
for acceptance and rejection of such services will be 
explored with participants. 

2. Dr Jonathan Sterne/Prof George Davey Smith: Causal 
Modelling for Life Course Epidemiology Standard statistical 
methods used in modelling epidemiological associations have 
limitations when used to assess the long-term effect of 
exposures measured on a number of occasions. Measurement 
error leads to underestimation of the size of associations, 
and proposed methods for correcting for such regression 
dilution bias depend on assumptions which may not be 
justified. The correlation between repeat measures means 
that assessing the relative importance of different 
exposure periods may be difficult, for example because of 
collinearity. The interplay between different exposures and 
outcomes over time may be complex; leading to 
"time-dependent confounding"; when the value of an exposure 
predicts the subsequent value of another exposure, and both 
are associated with the outcome. The purpose of this 
project is to use recently developed statistical approaches 
to improve on the methods currently used in life course 
epidemiology. The project will use a rich data set arising 
from the Caerphilly cohort, a cohort of originally middle 
aged men followed up in South Wales every 5 years for 20 
years. The study was originally set up to examine the 
epidemiology of coronary heart disease but its scope has 
been broadened to include a wider range of health outcomes. 
Statistical approaches to be evaluated include the use of 
structural equation models and graphical models fitted 
using Bayesian methods. These can explicitly incorporate 
measurement error and correlations between exposures over 
time. "Causal" models, introduced by Robins, will be used 
to control time-varying confounding.

3. Dr Joanna Coast: The impact and cost of antimicrobial 
resistance Antimicrobial resistance is an important social 
externality that has not been captured at the level of 
economic appraisal.  Although this is unsurprising given 
the difficulties associated with identifying and then 
costing the, often diffuse, impacts of resistance, it means 
that little information is available about the nature or 
extent of these costs.  Such information is of prime 
importance for assessing the relative value of alternative 
interventions intended to prevent and control the extent of 
resistance. This project will focus on developing methods 
for measuring the impact, and costs, associated with 
resistance to antimicrobials.  The student will need to 
review literature on the theory and conduct of economic 
evaluation, and the measurement and valuation of external 
effects, both positive and negative.  They will also need 
to develop a full understanding of the nature of 
antimicrobial resistance in order to identify where the 
effects of resistance are most likely to impact.  From this 
knowledge, methods for examining in detail the costs 
associated with resistance will be developed, incorporating 
both the identification of relevant impacts and the 
development of research tools to measure these impacts.  
Empirical case studies will be used to test and refine 
these methods.  This multi-disciplinary research will give 
the student a detailed understanding of antimicrobial 
resistance and the economics of externality, but at a 
broader level the student will also develop a critical 
understanding of epidemiological and economic research 
design and will gain experience in reviewing economic and 
epidemiological publications.


For an information pack about research studentships 
and contact details for advisers, apply to Mrs Julia 
Mackay, Department of Social Medicine, University of 
Bristol, Canynge Hall, Whiteladies Road, Bristol BS8 2PR 
(0117 928 7337), [log in to unmask]
Deadline for applications: 30th April 2000.









%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%